9 months into an AI project: data infrastructure is the hard, cultural part
adrianscottcom · x · 2026-09-28
An investor (PadraicMcC) shared observations from 9 months into an AI project (about 4 more to go), built on Azure + Databricks + Anthropic with a dedicated Nimble Gravity engineering team. A self-described tech skeptic, his core test is whether it helps close more deals, and he believes differentiation lies in sourcing and building.
Key takeaways:
- If your firm's data is poorly organized or lacks people/process/tool infrastructure around data management, fix that first — it's a heavy lift and more cultural than technical.
- Quoter adrianscottcom adds: any project should deliver at least one no-brainer win within 3, max 6 months.
More from Companies & People
- MIT Lab Teams With e14fund for ScienceClaw Hackathon on Collective AI for Science — ProfBuehlerMIT · 2026-09-28
- Opinion: Is the AI Safety Wave Really a Marketing Bet? — ro_ma_ro · 2026-09-28
- FTX's unliquidated portfolio would be worth ~$206B today, with Anthropic up 340x — panickssery · 2026-09-28
- X's invite-only bot creator program raises calls for agent usage dashboards — RachelVT42 · 2026-09-28
- OpenAI's Logan Kilpatrick: revenue-generating agent companies arrive at scale in 2027 — OfficialLoganK · 2026-09-28
- OpenAI DevDay to showcase persistent-memory agent, continual learning and io hardware — VraserX · 2026-09-28